Identifying Medicare Beneficiaries With Delirium

Area of Science:

• Geriatric Medicine
• Health Informatics
• Data Science

Background:

• Delirium is a common, serious, and preventable condition affecting thousands of older adults annually.
• Current methods lack validated approaches for identifying delirium in large datasets like Medicare claims.
• Accurate identification is crucial for timely intervention and improved patient outcomes.

Purpose of the Study:

• To develop and evaluate classification algorithms for identifying delirium using longitudinal Medicare administrative data.
• To assess the performance of these algorithms against a gold standard adjudicated from electronic health records.
• To establish a reliable method for detecting delirium in large-scale health data.

Main Methods:

• A linked electronic health record (EHR)-Medicare claims dataset was utilized.
• A stratified random sample of 1002 patients underwent standardized EHR review by neurologists and psychiatrists to adjudicate delirium status.
• Classification algorithms were developed using Medicare claims data, including International Classification of Diseases, 10th Edition (ICD-10) codes, and performance was assessed via 10-fold cross-validation.

Main Results:

• The study included a cohort with a mean age of 75 years; 6% of patients had delirium.
• A simple classification model incorporating delirium diagnoses counts, age, dementia status, and antipsychotic medication receipt demonstrated superior performance.
• This model achieved a cross-validation area under the receiver operating characteristic curve of 0.88 (95% CI: 0.84-0.91).

Conclusions:

• A delirium classification model leveraging Medicare administrative data and ICD-10 codes can effectively identify beneficiaries with delirium.
• This approach enables the detection of delirium in large datasets, facilitating broader public health surveillance and clinical management.
• The validated model offers a practical tool for identifying at-risk populations and improving care for older adults.